This research note explores the suitability of available datasets to analyse undercoverage and non-take-up (NTU) of minimum income (MI) schemes. It first reviews the characteristics of an ideal dataset that would allow researchers to properly estimate undercoverage and NTU, then describes two case studies for Belgium and Italy. The comparison between these two countries is interesting: Belgium has a long tradition of MI schemes, while Italy introduced national MI schemes only in 2018 (Inclusion Income, Reddito di Inclusione – REI, replaced in 2019 by the Citizenship Income, Reddito di Cittadinanza – RdC). The first case study looks at the undercoverage of MI schemes in both countries, using the (standardised) information provided in the European Union Survey on Income and Living Conditions (EU-SILC) (matched with administrative information in the Belgian case). This will measure the number of individuals/households satisfying the income test but not receiving the benefit. The second case study estimates NTU rates in Italy, using an innovative dataset that matches the Italian Household Budget Survey (HBS) with administrative data on RdC recipients. Both exercises also focus on the characteristics of individuals/households more likely to suffer from undercoverage or NTU of MI schemes.
The next section of this research note presents the most common and general data issues for the estimation of NTU and undercoverage. Two dedicated subsections deal with country-specific data is-sues for Belgium and Italy. The subsequent section presents the empirical exercises for the two coun-try case studies, while a final section summarises the main results, draws conclusions and provides some suggestions on improving the availability of data suitable to estimate NTU.

Disponível na